The Reproducibility of a Clinical Grading System of Motor Control in Patients with Low Back Pain

Objective: Over the past decade, instrument and palpation methods for quantifying the activation and recruitment of the transversus abdominis and lumbar multifidus have been proposed. Palpation methods however have recently been described and therefore have been subjected to little evaluation. One such palpation method is the Wisbey-Roth grading system. The recruitment of the transversus abdominis and lumbar multifidus is assessed in a series of functional body positions and movements. The ability to recruit these muscles is quantified by assigning 1 of 6 defined grades. The purpose of this study was to investigate the reproducibility of this grading system. Methods: A total of 2 meetings and 3 pilot trials were held with raters before commencement of the study to establish an agreed grading system protocol. Interrater reproducibility was investigated using a Latin square repeated measures design. Thirty-four subjects (62% male and 38% female; age range, 15-70 years) with a history of low back pain participated. A total of 4 practicing physiotherapists and 1 sports medicine physician graded subjects using the Wisbey-Roth grading system protocol. Results: Pair-wise weighted κ values ranged from -0.01 (95% confidence interval [CI], -0.33 to 0.31) to 0.56 (95% CI, 0.25 to 0.87), with average weighted κ being 0.29. The intraclass correlation coefficient (2,1) was 0.30 (95% CI, 0.15 to 0.48), and the standard error of the measurement was 1.6 units. Conclusions: The Wisbey-Roth grading system shows fair to poor reproducibility between raters. Therefore, it should not be used to exchange meaningful information between clinicians. Recommendations are made for further research and toward improving its reproducibility.